The timescale and mechanisms of fault sealing and water‐rock interaction after an earthquake

Abstract

Hydrogeochemical monitoring of a basalt-hosted aquifer, which contains Ice Age meteoric water and is situated at 1220 m below sea level in the Tjö rnes Fracture Zone, northern Iceland, has been ongoing since July 2002. Based on hydrogeochemical changes following an earthquake of magnitude (M w ) 5.8 on 16 September 2002, we constrained the timescales of post-seismic fault sealing and water-rock interaction. We interpret that the earthquake ruptured a hydrological barrier, permitting a rapid influx of chemically and isotopically distinct Ice Age meteoric water from a second aquifer. During the two subsequent years, we monitored a chemical and isotopic recovery towards pre-earthquake aquifer compositions, which we interpret to have been mainly facilitated by fault-sealing processes. This recovery was interrupted in November 2004 by a second rupturing event, which was probably induced by two minor earthquakes and which reopened the pathway to the second aquifer. We conclude that the timescale of fault sealing was approximately 2 years and that the approach to isotopic equilibrium (from global meteoric water line) was approximately 18% after >10 4 years.

Hydrothermal fluid flow within a tectonically active rift‐ridge transform junction: Tjörnes Fracture Zone, Iceland

“…The faults in the upflow zones are already partially open and probably mechanically weak. Pulses of overpressured fluids from the crystalline basement can move upwards along these faults during coseismic permeability enhancement, temporarily changing the fluid temperature and chemistry in the sedimentary basins as observed by Claesson et al [2004, 2007]. Notably, our numerical simulations as well as geological observations suggest an upflow zone of hot hydrothermal fluids in the vicinity of Húsavik.…”

“…If the flow system is disturbed during a seismic event such that the two fluid reservoirs begin to communicate, mixing of shallow and deep fluids is probable. This was proposed for the TFZ by Claesson et al [2004, 2007] who provided hydrogeochemical evidence of deep fluids rising to 1 km depth near the town of Húsavik immediately after a 5.8 M earthquake which occurred in September 2002 in the TFZ.…”

“…The black triangles are the locations of the boreholes and include the borehole “Húsavik‐Hola nr. 1”, where hydrogeochemical variations before and after a 5.8 M earthquake were observed [ Claesson et al , 2004, 2007]. …”

Hydrological effects of theMW7.1 Darfield (Canterbury) earthquake, 4 September 2010, New Zealand

“…The emergence of springs and the vast quantity of liquefied sand and silt carried to the ground surface after the earthquakes are indicators of the subsurface change to the aquifer system. Studies elsewhere have suggested that the impacts of earthquakes on hydrogeological systems may be relatively short-lived (e.g., Geballe et al 2011) although effects may last many months to a few years or more (e.g., Elkhoury et al 2006;Claesson et al 2007;Kitagawa et al 2007;Manga & Rowland 2009;Rudolph & Manga 2010;Wang et al 2012). The multiple earthquake responses recorded by DL/545 at Clyde appear to have groundwater-level offset and recovery times in the order of 200Á 300 d. At this stage, one year after the M W 7.1 Darfield earthquake, it is difficult to know the extent to which the hydrological impacts of the Canterbury earthquakes will be significant and/or long-term, although the preliminary eigen modelling motivates the hypothesis that there are certainly places in the near-field region experiencing significant longer term changes.…”

A Combined Ingress-Egress Model for the Kianna Unconformity-Related Uranium Deposit, Shea Creek Project, Athabasca Basin, Canada

“…Based on equations from Crank (1975) and diffusion constants from Cole and Ohmoto (1986), muscovite grains would have to have been in isotopic equilibrium with the >300 o C fluid for at least one million years in order for the oxygen and hydrogen stable isotopic compositions to be overprinted. This would mean that the fault structures remained open to heated fluids for at least one million years, which is unlikely, as faults generally seal immediately after rupture (Claesson et al, 2007;Kame et al, 2014).…”